Recently, gas separation based on the use of a membrane has been spotlighted as separation technology growing rapidly in response to the importance thereof. Gas separation using such a membrane has many advantages, including low energy consumption and operation cost and high processing utility, as compared to the conventional separation processes. Particularly, since 1980's, many fundamental studies using organic polymer membranes have been conducted. However, such conventional polymers generally impart efficient packing in a polymer chain space with little micropores present therein, and thus show relatively low mass transferability.
On the other hand, polymers having a high degree of free volume and known as microporous organic polymers have improved diffusibility in addition to adsorbability to small gas molecules, and thus have given many attentions as one of the prominent candidates useful for separation processes. Therefore, active studies have been conducted to develop organic polymers that can be used as gas separation membranes by focusing on the fact that a specific microporous polymer based on a rigid ladder-shaped structure having a twisted region inhibiting efficient packing in a polymer chain space provides relatively high gas permeability and selectivity.
Among such studies, many attempts have been made to apply rigid vitrified proaromatic organic polymers, such as polybenzoxazole, polybenzimidazole or polybenzthiazole, having excellent thermal, mechanical and chemical properties as gas separation membranes. However, most of the above organic polymers are hardly soluble in general organic solvents, and thus have difficulty in preparing membranes through a simple and practical solvent casting process. Under these circumstances, to overcome such difficulty, the inventors of the present invention have recently reported that a polybenzoxazole membrane obtained through a thermal rearrangement process of polyimide having a hydroxyl group at the ortho position provides a carbon dioxide permeability 10-100 times higher than the carbon dioxide permeability of the conventional polybenzoxazole membrane obtained by a solvent casting process. However, in this case, there is still a room for improvement in that the carbon dioxide/methane (CO2/CH4) selectivity is equal to that of the commercially available cellulose acetate membrane (Non-patent Document 1).
In order to improve the selectivity of a polybenzoxazole membrane, it has been also reported that a polybenzoxazole membrane obtained by thermal rearrangement of a membrane of a ortho-hydroxyl group-containing polyimide/poly(styrenesulfonic acid) blend at 300-650° C. provides a carbon dioxide/methane (CO2/CH4) selectivity improved by at most 95% as compared to a polybenzoxazole membrane obtained by thermal rearrangement of hydroxypolyimide containing no poly(styrenesulfonic acid). However, in this case, there is no disclosure about a method for preparing a polyimide used as a precursor for the preparation of the polybenzoxazole membrane. Thus, there is a problem in that no consideration is made about variations in the free volume factor and gas separation quality of a polybenzoxazole membrane rearranged thermally from a polyimide precursor depending on imidization methods of hydroxypolyimide, i.e., solution thermal imidization, azeotropic thermal imidization, solid state thermal imidization and chemical imidization (Patent Document 1).
Thus, based on the fact that the properties of a thermally arranged polybenzoxazole is affected by the method for preparing an aromatic polyimide, it has been reported that polybenzoxazole membranes are obtained by providing ortho-hydroxyl group-containing polyimides through various processes, such as solution thermal imidization, solid state thermal imidization and chemical imidization, and then carrying out thermal rearrangement thereof. However, the resultant membranes having high separation quality due to a specific porous structure derived from thermal rearrangement may be applied merely to separation membranes for removing water from ethanol or other organic solvents. Moreover, there is no suggestion about the quality as gas separation membranes (Patent Document 2).
Further, it has been reported that a crosslinked polybenzoxazole membrane obtained by providing ortho-hydroxyl group-containing polyimide through chemical imidization, carrying out thermal rearrangement thereof to obtain a polybenzoxazole membrane, and then subjecting the membrane to UV irradiation shows improved selectivity. However, in this case, since polyimide is obtained through chemical imidization, thermal imidization is omitted and the polybenzoxazole membrane rearranged thermally from the polyimide still has relatively low carbon dioxide permeability even though it has a crosslinked structure. Moreover, a UV irradiation system is required to form such a crosslinked structure, resulting in degradation of processability (Patent document 3).
Under these circumstances, the inventors of the present invention have focused on the fact that the gas transport behavior of a thermally rearranged polybenzoxazole depends on the imidization method of its precursor, polyimide, and the crosslinked structure of polybenzoxazole. Thus, the inventors have thought that a crosslinked polybenzoxazole membrane obtained by providing hydroxypolyimide having a hydroxyl group at the ortho position through solution thermal imidization, subjecting the resultant hydroxypolyimide to chemical crosslinking to form a crosslinked hydroxypolyimide membrane before the thermal rearrangement for forming polybenzoxazole, and then finally carrying out thermal rearrangement can improve the separation quality as a gas separation membrane significantly. The present invention is based on this thought.